转自:http://www.cnblogs.com/zhanglei644213943/archive/2009/12/27/1633356.html
【各大软件公司.NET工程师笔试压轴题】------不得不看
纵览各大社区、论坛,各大ORM框架火得不行了,如NHibernate、LINQ to SQL、ADO.NET Entity framework等,还有最近市场上出版的一本叫《领域驱动设计与模式实战》,里面也凸显了不少NHibernate在领域驱动设计中的作用与地位,也算是第一本与NHibernate相关的书籍吧!不过就NHibernate而言还是没有官方文档介绍得详细呵呵,园子里Kiler已经把他翻译成中文版的了,收益一大片仅仅是CET-4的人。不管你是用NHibernate也好,还是用LINQ to SQL也好,用profiler一跟踪,执行的都是SQL语句,所以所SQL是根。特别是对于那些以数据为中心的应用系统,在数据库中实现复杂的存储过程,复杂的报表查询,还是直接SQL来得痛快。当然对于那些在基于.NET的中间层应用中,它们实现面向对象的业务模型和商业逻辑的应用,NHibernate是最有用的。不管怎样,NHibernate一定可以帮助你消除或者包装那些针对特定厂商的SQL代码,并且帮你把结果集从表格式的表示形式转换到一系列的对象去(官方文档)。
有点跑题了,不再啰嗦----直接晾出压轴题。
压轴题第一问
1.把表一转换为表二
表一:
![](https://images.cnblogs.com/cnblogs_com/zhanglei644213943/1.png)
表二:
![](https://images.cnblogs.com/cnblogs_com/zhanglei644213943/佛挡杀佛.png)
数据库代码如下:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 DROP table #student
2 CREATE TABLE #student (stdname nvarchar(10),stdsubject nvarchar(10),result int)
3 INSERT INTO #student VALUES ('张三','语文',80)
4 INSERT INTO #student values ('张三','数学',90)
5 INSERT INTO #student VALUES ('张三','物理',85)
6 INSERT INTO #student VALUES ('李四','语文',85)
7 INSERT INTO #student values ('李四','数学',92)
8 INSERT INTO #student VALUES ('李四','物理',82)
9 INSERT INTO #student VALUES ('李四','化学',82)
10 INSERT INTO #student VALUES ('李四','化学',82)
11 SELECT * FROM #student
可能很多老手们,一看到这题目就有了答案。当然,贴出答案来不是我的目的,我要带着SQL新手们重构到答案。用李建忠老师最爱说的话就是------我不建议一上来就套用模式,而应该从重构到模式。
首先大家会想到分两组
1 select stdname,····,from #student group by stdname
然后······中间该写什么呢?
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 case stdsubject when '化学' then Result end
2 case stdsubject when '语文' then Result end
3 case stdsubject when '···' then Result end
4 case stdsubject when '···' then Result end
5 case stdsubject when '···' then Result end
表二里面得0是哪里来的呢?
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 isnull(sum(case stdsubject when '化学' then Result end),0)
2 isnull(sum(case stdsubject when '语文' then Result end),0)
3 isnull(sum(case stdsubject when '···' then Result end),0)
4 isnull(sum(case stdsubject when '···' then Result end),0)
5 isnull(sum(case stdsubject when '···' then Result end),0)
所以得出:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 select stdname,
2 isnull(sum(case stdsubject when '化学' then Result end),0) [化学],
3 isnull(sum(case stdsubject when '数学' then Result end),0) [数学],
4 isnull(sum(case stdsubject when '物理' then Result end),0) [物理],
5 isnull(sum(case stdsubject when '语文' then Result end),0) [语文]
6 from #student
7 group by stdname
然后得出答案:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 declare @sql varchar(4000)
2 set @sql = 'select stdname'
3 select @sql = @sql + ',isnull(sum(case stdsubject when '''+stdsubject+''' then Result end),0) ['+stdsubject+']'
4 from (select distinct stdsubject from #student) as a
5 select @sql = @sql+' from #student group by stdname'
6 print @sql
7 exec(@sql)
8
压轴题第二问:把表二转化为表一
表一:
![](https://images.cnblogs.com/cnblogs_com/zhanglei644213943/1.png)
表二:
![](https://images.cnblogs.com/cnblogs_com/zhanglei644213943/佛挡杀佛.png)
数据库代码如下:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 DROP table #student2
2 CREATE TABLE #student2 (stdname nvarchar(10),化学 int,数学 int,物理 int ,语文 int )
3 INSERT INTO #student2 VALUES ('李四',164,92,82,85)
4 INSERT INTO #student2 VALUES ('张三',0,90,85,80)
5 SELECT * FROM #student2
看到这题,直接想到:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 SELECT'李四'as stdname,stdname='化学', 化学 as result from #student2 where stdname='李四'
2 union all
3 SELECT'李四'as stdname,stdname='数学', 数学 as result from #student2 where stdname='李四'
4 union all
5 SELECT'李四'as stdname,stdname='物理', 物理 as result from #student2 where stdname='李四'
6 union all
7 SELECT'李四'as stdname,stdname='语文', 语文 as result from #student2 where stdname='李四'
8 union all
9 SELECT'张三'as stdname,stdname='化学', 化学 as result from #student2 where stdname='张三'
10 union all
11 SELECT'张三'as stdname,stdname='数学', 数学 as result from #student2 where stdname='张三'
12 union all
13 SELECT'张三'as stdname,stdname='物理', 物理 as result from #student2 where stdname='张三'
14 union all
15 SELECT'张三'as stdname,stdname='语文', 语文 as result from #student2 where stdname='张三'
重构到:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 declare @sql2 varchar(4000)
2 set @sql2 = ''
3 SELECT @sql2=@sql2+
4 'SELECT'''+stdname+'''as stdname,stdname=''化学'', 化学 as result from #student2 where stdname='''+stdname+'''
5 union all
6 SELECT'''+stdname+'''as stdname,stdname=''数学'', 数学 as result from #student2 where stdname='''+stdname+'''
7 union all
8 SELECT'''+stdname+'''as stdname,stdname=''物理'', 物理 as result from #student2 where stdname='''+stdname+'''
9 union all
10 SELECT'''+stdname+'''as stdname,stdname=''语文'', 语文 as result from #student2 where stdname='''+stdname+''' union all '
11 from (SELECT stdname FROM #student2) as a
12 SELECT @sql2 = LEFT(@sql2,LEN(@sql2) - 10)
13 PRINT(@sql2)
14 exec(@sql2)
如果要求不能出现 化学 数学 物理 语文 这样的关键字,那么可以这样写:
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
代码
1 select [name] into #tmpCloumns
2 from tempdb.dbo.syscolumns
3 where id=object_id('tempdb.dbo.#student2')
4 and [name]<>'stdname'
5 select * from #tmpCloumns
6
7 declare @strSql nvarchar(800)
8 select @strSql=''
9 select @strSql=@strSql+'union all'+char(10)+char(13)+
10 'select [stdname],'''+[name]+''' as [科目],['+[name]+']'+char(10)+char(13)+
11 'from [#student2]'+char(10)+char(13)
12 from #tmpCloumns
13
14 select @strSql=substring(@strSql,11,len(@strSql))+'order by stdname,[科目]'
15 --print @strSql
16 exec(@strsql)
这种题目,在各种笔试中出现的概率还是非常大的,大家不用死记。以前有的朋友看着复杂的报表查询,几百行SQL,望而生畏,然后说:"这是哪个SQL超人写的啊!"其实,谁一上来不可能写出那么长的SQL,也是慢慢重构--调试--重构-······